Also try force closing amber and opening it again of it doesn't work with another browser
Discussion
Nothing useful on the logs (at least for my yes). Not working with Chrome, not working after force stopping Amber, not working after fully removing it and installing a fresh APK. And again, with nak it works (including on Firefox, with all clients and all relays)
Also, now that I installed the fresh APK I can't log in on Amethyst either, it loads Amber and gets stuck in this screen, even after I force stop both apps.
What device are you using?
This is a Samsung S24 Ultra, bog standard normie phone. No VPNs, no Tor, as vanilla as it gets.
Always a Samsung or Xiaomi when those crazy bugs happens 🤣🤣
Have you also tried rebooting the device?
Just tried in a Samsung and it worked I have no idea what's happening with your device
Will do. Seems like I can load your message on Amethyst but can't DM back. And 0xChat is crashing on me, but I'll try it and reply back to you here.
I receive your messages. It looks like amethyst is sending then and not showing the note or any indicator that it still trying to send the message nostr:nprofile1qqsyvrp9u6p0mfur9dfdru3d853tx9mdjuhkphxuxgfwmryja7zsvhqpzamhxue69uhhv6t5daezumn0wd68yvfwvdhk6tcpz9mhxue69uhkummnw3ezuamfdejj7qgswaehxw309ahx7um5wghx6mmd9u2mk7fe
Yes, Amethyst is getting stuck over and over again, even when displaying events from my timeline.
So, restarting didn't work, downgrading to 4.0.4 worked (sorta).
I can login using the bunker flow and Citrine, it pops up an error about "mute kind something something" but eventually it works.
https://haven.accioly.social/d3d7328693e477f8c03e418a6a431a0a8d0b481cb0c7cd115369c71878edad25.mp4
But even 4.0.4 isn't letting me login with a standard nostrconnect flow which is the only option in some clients out here.
https://haven.accioly.social/9e2f6067c536d147c04a62c75a240c104d8b6b40bf1840bda278caf7735f2958.mp4
With 4.0.4 I can also log in to Amethyst.
IMO there's definitely a regression here, or rather several, as Amber used to work even with nostrconnect flows for me. Unfortunately I don't have another device to test this. What do you want to do here? Happy to drop this as I'm not even really using Nostr that much anymore, but if you want me to do anything to help you troubleshoot this, happy to DM you logs, grant you remote access or do any other tests you require.
I sent you a test apk in a dm. Did you receive it?
Basically I just updated the dependencies and tried to fix the app not closing when you accept or reject a permission
Not sure if it will fix the connection issue in this apk
Let me know if you received it and if it fixed something
So, with this version I can log in using Citrine and the bunker url flow, but still can't with the nostrconnect flow. Amber window now closes, but it seems like things get sorta stuck after that, I.e. clients keep struggling to load timelines, etc.
Ok, if I kill Amethyst, Jumble and Nosotros gets unstuck, so maybe Amethyst is hoarding all of Amber's capacity. Still can't log in with thr nostrconnect flow though.
Also, on jumble I'm still getting: "mute: no one was listening to your ephemeral event and it wasn't handled in any way, it was ignored"
That could be it, I think amethyst is using all the resources and the other apps can't use it. I'll try to fix this
The older versions of amber works right?
4.0.4 works to login with Citrine + a Amber generated Bunker URL, I still haven't found a version that works for me for nostrconnect URL flows. But it used to for sure (can't say which version exactly as I haven't updated Amber for a while)
in jumble you click the J and amber opens, just a silly thing to say, are you doing this and it's not working?
Yes, this is the nostrconnect (client initiate) flow. Clicking and copying and pasting the nostrconnect token is ultimately the same thing. Neither work for me unfortunately (the flow where I first create a bunkerurl in Amber works)
In nak works because you don't change apps when you connect to it. When you open amber the browser suspends the relay connection and then unsuspend it when you open the page again
If there's no failed to send logs or relay disconnected logs amber is working just fine
Relays should also implement the mute keyword fiatjaf added some months ago so we can always be sure when a client is losing the messages
How is it working just fine if I can't login to any application using any client in any browser? 🤣
And using any intermediate relay... Including Citrine on the same device...
one way to tell where the big is, .. do a nak stream of those events while you sign-in, you should see them flowing into the relay from amber. if you don't see them, then it's a bug in amber. if you *do see them, then it's a bug in the app..
that's my technique anyway.. 😁
This one seems very likely to be a bug on Amber by elimination (i.e. everything works together as long as Amber isn’t in the flow). Nak can act as a bunker on its own, but it’s a major hassle to use it on mobile, as it means having my VPN on and SSHing into my linux box to retrieve bunker URLs when I’m outside my network. Plus, running Amber and Citrine locally is the only way I know of for remote signing to work even relatively well, as NIP-46 is extremely sensitive to latency. Even the latency of a WireGuard VPN is enough to degrade the experience, with Amethyst becoming absolutely sluggish, almost unusable.
well, it's also true what greenart is sayiny is that when app switching, connections tend to get all screwed up (in the browser or maybe amber too, thats why watching the events is helpful, just assume, its not the relays because you tried multiple known working relays aready). I was having some random android voodoo issues this morning with meshtastic and I installed the pending OS update/rebooted, and they went away..
didja reboot? 😁
Aye. Android background processing is a pain and nostr pool connection management is broken enough by itself. Restarting Androix didnt work but nostr:nprofile1qqs827g8dkd07zjvlhh60csytujgd3l9mz7x807xk3fewge7rwlukxgppemhxue69uhkummn9ekx7mp0qydhwumn8ghj7argv4nx7un9wd6zumn0wd68yvfwvdhk6tcpz9mhxue69uhkummnw3ezumrpdejz7d9xcpk managed to release a test version that partially fixed things for bunkerurl flows (as long I don't use nostrconnect URLs and rawdog my nsec into Amethyst so that it doesn't hoard Amber resources things are working, or rather, back at acceptable levels of brokeness :()